Glycosylphosphatidylinositol (GPI)-anchored proteins (GPI-APs) are ubiquitous in eukaryotes. Examples of GPI-APs include folate receptor, acetylcholinesterase, renal dipeptidase and the variant surface glycoproteins of Trypanosoma brucei, the causative agent of African sleeping sickness. Inability to synthesize GPI-APs results in embryonic lethality in mammals. Defective GPI biosynthesis in multipotent hematopoietic human stem cells causes paroxysmal nocturnal hemoglobinuria, an acquired hemolytic disease. GPI-APs are needed for fungal cell viability and they are important in diseases such as trypanosomiasis, malaria and leishmaniasis that are caused by parasitic protozoa. The GPI assembly pathway is a drug target for fungal and protozoal diseases. GPI anchoring is catalyzed by GPI transamidase (GPIT), a 5-subunit membrane- bound complex located in the endoplasmic reticulum (ER). The catalytic subunit, Gpi8, shares homology with caspases; the functional role of the other subunits is unclear, but all are required for GPIT activity. Three of the non-catalytic subunits are over-expressed in certain cancers, indicating a link between GPIT and oncogenesis. In this R21 application we propose to initiate structure-function studies of the GPIT complex using electron microscopy and X-ray crystallography. We are ultimately interested in establishing the structural organization of GPIT, delineating the role of its subunits, and understanding how this important enzyme is regulated. In two specific aims we propose to (1) analyze the endogenous GPIT complex from yeast by electron microscopy and (2) express GPIT subunits and sub-complexes for X-ray crystallographic studies. Our efforts are expected to yield a medium-resolution structure of GPIT and pave the way for a future high-resolution structure of the intact complex. These studies will have high impact as there is no structural information on GPIT; also, results obtained here will shed light on other multi-subunit membrane bound enzymes in the ER such as oligosaccharyltransferase and signal peptidase that play a critical role in processing a wide range of ER-translocated proteins, including proteins destined for GPI anchoring, but whose functional architecture remains largely a mystery.